Method and apparatus for computer system diagnostics using safepoints

Abstract

A method for safepointing a system that includes receiving a stop command by an executing thread from a master, wherein the executing thread executes an operating system, continuing execution of the executing thread until a safepoint is reached after receiving the stop command, halting execution of the executing thread at the safepoint; and evaluating a response from the executing thread to diagnosis the system.

1. A method for safepointing a system comprising:
 receiving a stop command by an executing thread from a master, wherein the executing thread executes an operating system; 
 continuing execution of the executing thread until a safepoint is reached after receiving the stop command; 
 halting execution of the executing thread at the safepoint; and 
 evaluating a response from the executing thread to diagnosis the system. 
 
     
     
       2. The method of  claim 1 , further comprising:
 inserting the safepoint in source code of the operating system. 
 
     
     
       3. The method of  claim 2 , wherein the safepoint comprises safepoint code inserted into source code executing on the system. 
     
     
       4. The method of  claim 3 , wherein safepoint code is inserted into source code using dynamic code generation. 
     
     
       5. The method of  claim 1 , wherein evaluating the response comprises detecting an error if a response is not received from the executing thread. 
     
     
       6. The method of  claim 5 , further comprising diagnosing the system using the error. 
     
     
       7. The method of  claim 1 , wherein the master comprises a plurality of master nodes. 
     
     
       8. The method of  claim 7 , wherein the plurality of master nodes synchronize sending the stop command. 
     
     
       9. The method of  claim 1 , wherein the frequency of the stop command is configurable. 
     
     
       10. The method of  claim 1 , further comprising:
 inserting the safepoint in an application; 
 receiving a stop command by an application thread from the master; 
 continuing execution of the application thread until a safepoint is reached after receiving the stop command; 
 halting execution of the application thread at the safepoint; and 
 evaluating a response from the application thread to diagnosis the system. 
 
     
     
       11. The method of  claim 1 , further comprising:
 receiving a restart command from the master by the executing thread if the response is received by each executing thread. 
 
     
     
       12. The method of  claim 1 , further comprising:
 forwarding the stop command from a node control master to a subordinate after reaching the safepoint. 
 
     
     
       13. The method of  claim 12 , further comprising:
 forwarding a restart command to the subordinate by the node control master. 
 
     
     
       14. The method of  claim 12 , wherein evaluating the response comprises detecting an error if a response is not received from the subordinate. 
     
     
       15. The method of  claim 14 , wherein the error is used to diagnosis the system. 
     
     
       16. A system for safepointing comprising:
 a plurality of subordinates configured to:
 receive a stop command, 
 continue execution until reaching a safepoint after receiving the stop command, 
 halt execution at the safepoint, and 
 forwarding a response if the stop command is processed; and 
 
 a master configured to:
 send the stop command to a subordinate in the plurality of subordinates, wherein the subordinate is an operating system thread; and 
 determine whether the response is received from each subordinate in the plurality of subordinates, 
 wherein the response from each subordinate is evaluated to diagnosis the system. 
 
 
     
     
       17. The system of  claim 16 , wherein the safepoint comprises safepoint code inserted into source code executing on the system. 
     
     
       18. The system of  claim 17 , wherein safepoint code is inserted into source code using dynamic code generation. 
     
     
       19. The system of  claim 16 , wherein the master is further configured to send a restart command to the subordinate if the response is received from each subordinate in the plurality of subordinates. 
     
     
       20. The system of  claim 16 , wherein evaluating the response comprises detecting an error if a response is not received from each subordinate. 
     
     
       21. The system of  claim 20 , wherein the error is used to diagnosis the system. 
     
     
       22. The system of  claim 16 , wherein the master comprises a plurality of master nodes. 
     
     
       23. The system of  claim 22 , wherein the plurality of master nodes synchronize sending the stop command. 
     
     
       24. The system of  claim 16 , wherein the frequency of the stop command is configurable. 
     
     
       25. The system of  claim 16 , further comprising:
 a node control master configured to:
 receive a stop command from the master, and 
 forward the stop command to each subordinate in the plurality of subordinates after reaching the safepoint. 
 
 
     
     
       26. The system of  claim 25 , wherein the node control master is further configured to:
 forward a restart command to each subordinate in the plurality of subordinates. 
 
     
     
       27. A distributed system having a plurality of nodes comprising:
 a plurality of subordinates configured to:
 receive a stop command, 
 continue execution until reaching a safepoint after receiving the stop command, 
 halt execution at the safepoint, and 
 forwarding a response if the stop command is processed; and 
 
 a master configured to:
 send the stop command to a subordinate in the plurality of subordinates, wherein the subordinate is an operating system thread; and 
 determine whether the response is received from each subordinate in the plurality of subordinates, 
 wherein the response from each subordinate is evaluated to diagnosis the system.